Download - Reciprocity in Interaction: A Window on the First Year of Life in Autism

Hindawi Publishing CorporationAutism Research and TreatmentVolume 2013 Article ID 705895 12 pageshttpdxdoiorg1011552013705895

Research ArticleReciprocity in Interaction A Window onthe First Year of Life in Autism

Fabio Apicella1 Natasha Chericoni1 Valeria Costanzo1 Sara Baldini1

Lucia Billeci2 David Cohen3 and Filippo Muratori14

1 Scientific Institute ldquoFondazione Stella Marisrdquo Viale del Tirreno 331 I-56018 Calambrone Pisa Italy2 Institute of Clinical Physiology National Council of Research (CNR) Via G Moruzzi 1 56124 Pisa Italy3 Department of Child and Adolescent Psychiatry Groupe Hospitalier Pitie-Salpetriere APHP Universite Pierre et Marie Curie47 bd de lrsquoHopital 75013 Paris France

4Department of Developmental Neuropsychiatry University of Pisa Viale del Tirreno 331 I-56018 Calambrone Pisa Italy

Correspondence should be addressed to Fabio Apicella fabioapicellainpeunipiit

Received 11 January 2013 Revised 26 March 2013 Accepted 17 April 2013

Academic Editor Bennett L Leventhal

Copyright copy 2013 Fabio Apicella et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

From early infancy onwards young children appear motivated to engage reciprocally with others and share psychological statesduring dyadic interactions Although poor reciprocity is one of the defining features of autism spectrum disorders (ASDs) fewstudies have focused on the direct assessment of real-life reciprocal behavior consequently our knowledge of the nature and thedevelopment of this core feature of autism is still limited In this study we describe the phenomenon of reciprocity in infant-caregiver interaction by analyzing family movies taken during the first year of life of 10 infants with ASD and 9 infants withtypical development (TD) We analyzed reciprocal behaviors by means of a coding scheme developed for this purpose (caregiver-infant reciprocity scale (CIRS)) Infants with ASD displayed less motor activity during the first semester and subsequently fewervocalizations compared to TD infants Caregivers of ASD infants showed in the second semester shorter periods of involvementand a reduction of affectionate touch These results suggest that from the first months of life a nonsynchronic motor-vocal patternmay interfere in different ways with the development of reciprocity in the primary relationship between infants later diagnosedwith ASD and their caregivers

1 Introduction

Reciprocity can be defined as an interactive condition inwhich two individuals mutually respond to each other whileperforming activities together [1] This attitude is presentfrom the first moments of life when an infant is naturallyorientated towards the motherrsquos face and responds preferen-tially to it Reciprocity is necessary in order to reach sharedgoals and consists of symmetrical exchanges characterizedby finely tuned turn-taking [2 3] From the first interactionswith their caregivers typically developing infants (TD) showa natural aptitude for engaging in joint action and for sharingpsychological states [4ndash6] Mutual exchanges exist from thebeginning when each member of the dyad is responsive tothe other trying to engage the other in turn and changinghisher behavior according to the otherrsquos solicitations [7]

Poor reciprocity is one of the defining features of autismspectrum disorders (ASDs) a class of neurodevelopmentaldisorders that disrupt regular interactions in the social realm[8] A widely shared hypothesis is that abnormalities in theprocessing of and in responding to stimuli coming fromthe social environment are in action from the first stages oflife with cascading effects on mutual exchanges in dyadicinteractions Despite the fact that a reduced reciprocity isfrequently reported as one of the earliest signs of autismfew studies have directly assessed reciprocity during infant-parent interactions in a naturalistic setting consequently thenature and the developmental course of this core feature inautism are still open issues

Some studies on early autism both retrospective andprospective have described behavioral difficulties thatare in some way linked with the concept of reciprocity

2 Autism Research and Treatment

Palomo et al [9] suggested that in infants with ASDcompared to TD infants there are some differences in theability to interact reciprocally because of a reduction inthe capacity of the former to engage with the other andto orient toward social stimuli Wan et al [10] found thatinfants at-risk of ASD showed lower levels of liveliness thatis a general tendency toward less activity during exchangesWe have described a general difficulty in maintaining socialengagement and in accepting invitation [11] Other studies[12 13] have revealed that during mutual exchange betweeninfant and parent while the former serves as the leadingmember in a typical dyad this does not happen in dyads withan infant with ASD Infants with these social difficulties offertheir parents who are in need of both responses and promptsfrom their infants fewer opportunities for interaction Whileinfantsrsquo social difficulties have been sufficiently analyzedand assessed their effects on parental attitude have receivedless attention Some studies have described longitudinalchanges in the intensity and quality of parental solicitationsSaint-Georges et al [14] focused on sequentially analyzedinteractions and they found that parents of infants with ASD(1) respond to their infants at the same level as parents ofinfants with TD but (2) show a more directive interactivestyle characterized by an increase in stimulating and touchingactivity Congruently Doussard-Roosevelt et al [15] andTrevarthen and Daniel [16] found that parents of childrenwith ASD overstimulate their infants as a result of theirinactivity Van Ijzendoorn et al [17] observed that parentsof children with ASD tended to answer their childrsquos signalswith more physical contact and fewer verbal approaches Wehave described [11] a reduction in caregiver behaviors aimedat downregulating the infantrsquos arousal and mood as one ofthe first expressions of lack of initiative in infants with ASD

This evidence of an early alteration in parent-infantinteraction suggests that the first signs of ASD could be betterunderstood by observing interactive dynamics rather thanthe increase or decrease in single behaviors of one memberof the dyad As stated by Thomas and Martin [18] and byAnderson et al [1] parent-infant reciprocity is better studiedby analyzing the effects on the other member of each dyadmemberrsquos behavior

On the basis of these assumptions we have hypothesizedthat reciprocity can be better understood by analyzing dyadicbehaviors that occur in simple caregiver-infant interactionsobserved in family home videos a valid method for observ-ing ecological interactions from the first stages of life Todescribe interactions specifically in terms of reciprocity wehave developed the caregiver-infant reciprocity scale (CIRS)which consists of items referring to mutual exchanges (suchas involvement or responsiveness) and single or molecularbehaviors (such as vocalization for infants or name promptfor caregivers) The CIRS was developed in order to describeand evaluate how infants try to involve and to respond to theotherrsquos bids according to the literature [9 11] we expectedin infants with ASD a reduced frequency of behaviors usedto involve and to maintain the other in interaction At thesame time the CIRS enabled us to explore the effect of theinfantrsquos behavior on parentsrsquo interactive attitudes accordingto the literature we expected some differences in terms of

type and quantity of stimulations but not in terms of globalavailability [17] The CIRS was also developed to explorewhether and to what extent abnormalities in the frequencyof motor behaviors contribute to disturbing the developmentof the atypical infantrsquos ability to engage in interactions ourhypothesis based on a previous exploratory study was thatabnormalities in motor activity can constitute a warningsignal for difficulties in reciprocity

2 Methods

21 Participants We studied retrospective home moviesfrom the first year of life of two groups of children Thefirst group was composed of 10 children (MF 91) with adiagnosis of autism spectrum disorder (ASD) performed bya child psychiatrist according to the DSM-IV-TR criteria [19]and confirmed by ADOS-G and ADI-R scores At the time ofthe diagnosis their ages ranged between 4 and 6 years Theywere also administered Griffithsmental developmental scale-ER in order to determine intellectual functioning (meanIQ 5926 sd 849) The second group was composed of9 typically developing (TD) children who were recruitedamong children attending a local kindergarten these childrendid not present any abnormal medical or developmentalcondition as confirmed by nonclinical scores using the childbehavior checklist [20] At the time of recruitment theTD children were aged between 4 and 5 years The videosused in the study were taken from a larger collection ofvideos held at the Stella Maris Institute University of PisaItaly

22 Video Collection and Editing Procedures Families wereasked to provide any videotape recorded during their childrsquosfirst year of life Copies of the videos were made and codedwith an ID number to preserve confidentiality Writteninformed consent was collected from participating familiesThe selection of the video clips was carried out by a researchassistant naive to the childrenrsquos diagnoses Only sequenceswhere the child was involved in an interaction with avisible caregiver were selected for the study Sequences wereselected in terms of situation type We included face-to-faceinteractions in various circumstances such as bath timenutrition time and play time We also included face-to-faceinteractions per se (ie the baby on the changing table and themother in front of himher talking in motherese) Birthdayparties and sequences with several adults andor childrensimultaneously present in the video were excluded

A t-test was performed to check that the selectedmaterialwas comparable across groups in terms of total durationand number and type of situations In accordance withother studies in the literature [11 21 22] which analyzedthe developmental pattern of infants across the first year oflife we considered the first and second six months of lifeseparately in order to take into account the multiple changeswhich take place during this period The total duration ofthe videos analyzed was 46 minutes for the age period 0ndash6months (hereinafter T1) and 51 minutes for the age period6ndash12 months (hereinafter T2) We analyzed a total of 81

Autism Research and Treatment 3

Table 1 Caregiver-infant rating scale

Infant behaviors

States(1) InvolvementA time interval in which the child tries to attract thecaregiverrsquos attention and to retain it Involvement starts whenthe child solicits an interaction and stops when the childshifts hisher attention toward an object and abandonshisher active role in the interaction possibly moving onto aresponsiveness state

For example the child starts to look and vocalizetoward the caregiver while heshe is talking to someoneelse in the room and not looking at the child

Responsiveness A time interval in which the child appears disposed torespond appropriately to the caregiverrsquos attempts to attracthisher attention Responsiveness starts when the childaccepts the caregiverrsquos invitation to participate in theinteraction and stops when heshe is no longer interested inthe other or when heshe shifts hisher gaze and directshisher attention away or takes an active role starting aninvolvement state

For example the child is looking at an object and thecaregiver calls hisher name looks at and smiles athimher in order to obtain hisher gaze and attentionThe child shifts hisher gaze towards the caregiver andstarts to vocalize towards himher

Events(2) Vocalization(a) of involvement(b) of responsiveness

The child vocalizes and looks toward the caregiverOnly vocalizations directed to the caregiver are consideredVocalizations directed toward objects or toward someonedifferent from the caregiver (undirected) are not codedVocalizations are distinguished according to their function inthe interaction(a) Vocalizations are used to start an interaction or to attractthe caregiverrsquos attention(b) Vocalizations are used to respond to the caregiverrsquossolicitations

Activating behavior(a) with gaze(b) with motor activity

The child provokes an interaction in order to involve thecaregiver For this purpose heshe can(a) orient their gaze toward the caregiver(b) perform motor activity toward the caregiver (ie movinglimbs touching grabbing reaching out and movingtoward) Motor activity is considered as an activatingbehavior only if it is used with the purpose of attracting thecaregiverrsquos attention Spontaneous movements are not coded

Attuned behavior(a) with gaze(b) with motor activity

The child responds to the caregiverrsquos attempts to attracthisher attention For this purpose heshe can(a) orient their gaze toward the caregiver when the lattersolicits hisher attention(b) perform motor activity directed towards the caregiver(ie adapting posture to caregiverrsquos body moving limbsreaching out etc) Motor activity is considered as an attunedbehavior only if it is used with the purpose of responding tothe caregiverrsquos stimulations Spontaneous movements are notcoded

Smile(a) Coordinated(b) Uncoordinated

The infant smiles in different ways The smile is distinguishedaccording to the direction of the gaze(a) The infantrsquos smile is coordinated with the direction ofhisher gaze towards the caregiver(b) The infantrsquos smile is not coordinated with hisher gazetowards the caregiver


Table 1 Continued

Caregiver behaviorsStates(1) Involvement A time interval in which the caregiver tries to attract the infantrsquos

attention and to retain it Involvement starts when the caregiver solicitsthe infant to participate in an interaction and stops when hesheabandons hisher active role (ie the caregiver directs hisher attentiontoward an object or abandons hisher active role in the interactionchanging to a responsiveness state)

For example while the child is looking at an object the caregivercalls hisher name in order to obtain hisher gaze and hisherattention or heshe looks or smiles at himher

Responsiveness A time interval in which the caregiver responds appropriately to thechildrsquos attempts to attract hisher attention and to retain itResponsiveness starts when the caregiver responds to the infantrsquosinvitation by participating in an interaction and stops when heshe is nolonger interested in the interaction or when heshe changes hisher State(ie when heshe changes to an involvement state)

For example the caregiver is talking with someone else inthe room and heshe is not looking at the child The child starts tolook at the caregiver and vocalizes toward himher in order toobtain hisher attention and the caregiver responds by looking atthe child and smiling or vocalizing toward himher


The caregiver vocalizes towards the infant looking at himherOnly vocalizations directed to the child are considered Vocalizationsdirected toward objects or toward someone different from the child(undirected) are not codedVocalizations are distinguished according to their function in theinteraction(a) Vocalizations are used to start an interaction or to attract the childrsquosattention(b) Vocalizations are used to respond to the childrsquos solicitations

Name prompt The caregiver calls the infant by name

Affectionate touch(a) of involvement(b) of responsiveness

The caregiver touches the infant in an affectionate way for examplecaressing or kissing himher Affectionate touch is also distinguishedaccording to its function in the interaction(a) Touch is used to start an interaction or to attract the childrsquos attention(b) Touch is used to respond to the childrsquos solicitations

Stimulating gesture(a) of involvement(b) of responsiveness

The caregiver gesticulates tickles makes faces or presents the infantwith objects Stimulating gestures are distinguished according to theirfunction in the interaction(a) Gestures are used to start an interaction or to attract the childrsquosattention(b) Gestures are used to respond to the childrsquos solicitations

(1)Duration and frequency are considered(2)Frequency is considered

interactive sequences for infants with ASD and 80 for TDinfants

23 Measures We developed the caregiver-infant reciprocityscale (CIRS) on the basis of previous research using theinfant caregiver behavior scale (ICBS) developed byMuratoriet al [11] and the emotional availability scales (EASs) byBiringen [23] We were interested in studying caregiver-infant interaction in terms of reciprocal involvement (iethe attempt to attract the other into interaction) and respon-siveness (ie the behavioral response to the otherrsquos attempt

to start an interaction) focusing on the specific behaviorswhich characterize these items rather than describing thesecapacities in terms of level (as does the EAS) The CIRS(see Table 1 for the description of the items) is composedof 12 items describing both caregiver and infant behaviorsused to respond to or involve the other (6 items refer tothe caregiver and 6 items refer to the infant) For both setsof items 2 (involvement and responsiveness) are evaluatedas states in that they represent the length of time spent oneither activating or responsive behaviors Next to these items4 other items referred to as events are used to describemore specifically the single behaviors that are enacted to


elicit a response from the other member of the dyad or torespond to the otherrsquos solicitations For these latter itemsonly the frequency and not the duration of each is recordedFor the purpose of our study we chose to code only thosebehaviors which were directed to the other member of thedyad and specified for each function in terms of involvementor responsiveness In this way vocalizations of the caregiverdirected to another adult or behaviors of the infant whichwere not related to the interaction with the caregiver werenot taken into account

24 Computer-Based Coding System The Observer XT 1001015840[24] a professional software for the collection managementand analysis of observational data was used This softwarehad already been used to study behavioral patterns in infantswith autism and its feasibility had been demonstrated [11]The Observer was configured for the application of the CIRSto our videotapes We have used the ldquoObserver Video-Prordquoa software from the Observer package which is able toread time information directly from the video allowing anaccurate event timing at the chosen playback speed

25 Codersrsquo Training and Intercoder Agreement ProceduresTwo coders (NC and VC) with an undergraduate degreein psychology were trained to use the coding system byobserving and coding videos of both TD infants and infantswith ASD who were not included in the study The two-month training period enabled the coders to become familiarwith the meaning of the items to identify them correctlyand to acquire ability in coding procedures In additionthe coders were required to achieve a satisfactory level ofagreement between them (Cohenrsquos Kappa ge 070) and withan expert clinician (FA) in two matched sequences for eachage range The tolerance window regarding time discrepancybetween coding was set at 1 s for event behaviors and at3 s for state behaviors The coding recorded outside thesewindows was reported as a coding error and was considereda disagreement In order to avoid any interpretation biasesthe two coders were blind to group membership Intercodersagreement was calculated for each item and the values of119896 were ge 070 The mean Intercoder reliability calculateddirectly by the Observer XT showed satisfactory agreement(119896 = 081) In order to verify the ongoing agreement 25 ofthe collection of sequences were coded by both coders Themean Intercoder agreement calculated on these sequenceswas k = 084 and the values of 119896 were ge 070 In all cases ofdiscordance between coders a third coderrsquos (FA) advice wasused

26 Data Analysis Due to the different length of videosequences of each child the frequencies and the durationof each behavior were converted respectively to a rationumber of behaviors per time (hereinafter ratemin) and toa percentage duration (hereinafter duration) calculatedbetween the total duration of a behavior and the durationof the sequences available for the same subject [11 21]Two additional items were further created for the purposeof analysis by merging activating behaviorsmdashwith motor

activity and attuned behaviorsmdashwith motor activity in aunique item named total motor activity and by mergingactivating behaviorsmdashwith gaze and attuned behaviorsmdashwithgaze in a unique item named total gaze in order to evaluateto what extent these behaviors are used independently fromtheir purpose (to involverespond to the other person)

Rates and percentages of all CIRS items were comparedusing Studentrsquos t-test for independent samples in order todetect ldquobetween grouprdquo differences (separately for T1 andT2) and Studentrsquos t-test for paired samples to detect ldquowithingrouprdquo differences between age periods (separately for ASDand TD) Statistical analysis was performed using SPSSversion 160 [25] Level of significance was set at 119875 lt 005

3 Results

Results are reported in Table 2 for infant behaviors and inTable 3 for caregiver behaviors

31 Group Comparison Both at T1 and at T2 there were nosignificant differences between groups in the ratemin and inthe percentage duration of the infant states (Table 2) At T1infants with ASD showed a significantly lower result for thefollowing events activating behaviormdashtotal and total motoractivity At T2 infants with ASD showed a significantly lowerresult for the event vocalizationsmdashtotal and in particularfewer vocalizationsmdashof responsiveness

As for caregiver behaviors (Table 3) both at T1 and at T2no significant differences were found between groups eitherfor states or events with the exception of the less affectionatetouchmdashof responsiveness in caregivers of infants with ASD atT1

32 Age Group Comparison In TD infants (Table 2) asignificant increase was found in the percentage durationof the state responsiveness A significant increase was alsofound in the ratemin of activating behaviorsmdashwith motoractivity attuned Behaviorsmdashwith motor activity and totalmotor activity Vocalizationsmdashtotal and vocalizationsmdashofresponsiveness showed a significant increase too Age groupcomparison did not show any significant differences in TDcaregivers either for states or events (Table 3)

In infants with ASD (Table 2) no significant differenceswere found for states in the transition between T1 and T2 Asignificant increase was found in activating behaviorsmdashtotalparticularly in activating behaviorsmdashwithmotor activity andin total motor activity In addition a significant decrease wasfound in vocalizationsmdashtotal

Caregivers of infants with ASD (Table 3) showed a sig-nificant decrease in the percentage duration of the stateinvolvement and in the ratemin of affectionate touch tak-ing into consideration both affectionate touchmdashTotalmdashandaffectionate Touchmdashof involvement

4 Discussion

This study described both segmental and longitudinal pecu-liarities in the context of the early developing relationship


Table 2 Infantrsquos behaviors means and standard deviations for ASD and TD 119905-test between groups at T1 and T2 and within groups in ASDand TD

Infantrsquos behaviorsT1(1) T2(2) 119905-test between groups 119905-test within groups119872 (SD) 119872 (SD) 119875 value(3) 119875 value(3)

ASD TD ASD TD T1(1) T2(2) ASD TDStates(4)

Involvement duration 301 (54) 83 (107) 58 (99) 91 (57) 018 039 041 085ratemin 03 (06) 06 (07) 05 (09) 09 (05) 032 025 048 019

Responsiveness duration 547 (26) 463 (23) 533 (301) 716 (171) 047 113 086 003lowast

ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

Activating behaviorsTotal 08 (11) 28 (28) 54 (51) 5 (31) 004lowast 081 002lowast 011Gaze 03 (04) 11 (13) 16 (21) 11 (12) 008 053 007 095Motor 05 (07) 18 (19) 38 (35) 39 (21) 007 099 002lowast 002lowast

Attuned behaviorsTotal 146 (78) 185 (65) 174 (106) 236 (111) 025 006 021 016Gaze 92 (46) 104 (51) 63 (57) 77 (63) 060 055 057 065Motor 53 (41) 81 (33) 71 (67) 145 (91) 013 023 024 005lowast

VocalizationsTotal 156 (116) 143 (91) 101 (102) 265 (111) 080 0004lowastlowast 003lowast 003lowast

Involvement 22 (4) 58 (8) 31 (43) 69 (38) 022 006 059 068Responsiveness 133 (11) 85 (63) 7 (81) 183 (125) 026 003lowast 007 002lowast

SmileTotal 76 (82) 58 (44) 43 (41) 74 (52) 056 017 019 037Coordinated 56 (72) 46 (31) 28 (29) 31 (22) 069 080 021 020Uncoordinated 11 (12) 07 (13) 08 (11) 16 (2) 056 024 051 029

Total Gaze(6) 95 (47) 115 (55) 119 (68) 102 (72) 040 061 034 069Total Motor(7) 59 (41) 98 (39) 109 (66) 184 (98) 004lowast 007 002lowast 002lowast(1)0ndash6 months period(2)6ndash12 months period(3)lowastlowast119875 le 001 lowastP le 005

(4)State the time interval in which the behavior is observed Duration and frequency are considered(5)Events single behaviors coded each time they are observed Frequency is considered(67)These two additional items were created merging motor and gaze behaviors allocated in the activating and attuned behaviors

between a small group of infants with ASD and their primarycaregivers Considering the small sample size it providesa provisional in-depth model for understanding the onsetand the perpetuation of deficits in social reciprocity Firstof all we found that while the time the dyads spent ininvolvement and responsiveness did not differ across groupsquite opposite trajectories emerged during the first year oflife suggesting that two different reciprocity profiles maybe under construction in infants with ASD compared toTD infants (see Figure 1) In contrast with the TD profilein fact the ASD profile is characterized by the absence ofa growth in the infantsrsquo responsiveness and by the paralleldecrease in the time their caregivers spend attracting theirattention in order to engage them in an interaction We

could hypothesize that differences in the parents and infantsrsquoprofiles are interconnected and that the parentrsquos level ofinvolvementmay bemodulated by the infantrsquos responsivenessand vice versa

Moreover the analysis of events suggests several qual-itative differences in the behaviors used by the infants inorder to involve or to respond to the caregiver First theearly significantly reduced amount of activating behaviors(total and with motor activity) and total motor behaviors ininfants with ASD highlights the lack ofmotor activity used bythese infants both to engage or to respond to caregiversrsquo bids(Figure 2) The reduced motor activity at T1 confirms otherstudies on home videos Adrien et al [26] observed higherlevels of hypotonia and the presence of unusual postures (see


Table 3 Caregiverrsquos behaviors means and standard deviations for ASD and TD 119905-test between groups at T1 and T2 and within groups inASD and TD

Caregiverrsquos behaviorsT1(1) T2(2) 119905-test between groups 119905-test within groups119872 (SD) 119872 (SD) 119875 value(3) 119875 value(3)

ASD TD ASD TD ASD TD ASD TDStates(4)

Involvement duration 87 (141) 769 (155) 666 (24) 824 (97) 015 008 004lowast 043ratemin 24 (11) 79 (112) 35 (73) 56 (65) 031 052 095 064

Responsiveness duration 37 (11) 3 (18) 19 (08) 22 (1) 040 056 041 023ratemin 03 (04) 06 (07) 03 (07) 04 (04) 032 079 087 057

Events(5)

Affectionate touchTotal 123 (77) 95 (81) 3 (26) 62 (74) 045 021 000lowastlowast 053Involvement 122 (76) 85 (74) 29 (24) 61 (73) 029 020 000lowastlowast 009Responsiveness 004 (01) 102 (14) 011 (02) 01 (02) 005lowast 082 051 042

VocalizationsTotal 633 (299) 555 (241) 475 (345) 665 (252) 054 069 017 054Involvement 592 (298) 491 (245) 444 (362) 565 (225) 044 040 022 097Responsiveness 41 (43) 64 (82) 31 (38) 64 (84) 044 027 060 034

Stimulating gesturesTotal 197 (109) 214 (131) 26 (137) 217 (111) 077 046 036 084Involvement 196 (11) 208 (132 248 (141) 197 (109) 082 039 044 012Responsiveness 01 (03) 05 (07) 12 (14) 19 (26) 016 041 007 095

Name prompt 27 (36) 24 (14) 31 (24) 26 (15) 082 055 072 083(1)0ndash6 months period(2)6ndash12 months period(3)lowastlowast119875 le 001 lowast119875 le 005

(4)State the time interval in which the behavior is observed Duration and frequency are considered(5)Events single behaviors coded each time they are observed Frequency is considered

also [27]) in infants with ASD Phagava et al [28] describeda very early pattern of poor motor repertoire in infants withASD Also retrospective interviews with parents of childrenwith ASD support hypotonia and reduced motor activityas early signs of ASD sometimes years before diagnosis isperformed [29] Some studies on siblings have describeddecreased activity at six months of life as a characteristic ofinfants who are developing an ASD [22 30] These data onyoung children are further supported by the meta-analysiscarried out by Fournier et al [31] on older children Wehypothesize that differences in motor activity during the firstyear of life and especially during the first 6 months maybe a key component in the understanding of the differentquality of reciprocity Further research could explore to whatextent this early lack of motor activity might jeopardizethe childrsquos ability to start and to respond to an interactiveexchangemaking the infant-parent interactionmore difficultand compromising the development of reciprocity

In the second six months the same activating and motorbehaviors significantly increase in infants with ASD as well asin TD infants although they remain at a slightly lower level inASD and attuned motor activity significantly increases onlyin TD infantsThis longitudinal data might suggest that even

if there is an improvement of motor activity in ASD thesechildren make less use of it to be attuned to others We couldhypothesize that the pattern composed of an early lower levelof motor activity and a subsequent nonincrease in attunedmotor activity could represent the expression or the substrateof ASD difficulties in reciprocity which in TD infants relieson the significant increase of attuned behaviorsmdashwith motoractivity

While in the first six months we only found differencesin items referring to motor activity during the second sixmonths of life lower rates in the amount of vocalizationsmdashtotal significantly distinguished ASD from TD (Figure 2)This difference was due to both the significant decreasein vocalizations in particular of the responsiveness typein the ASD group and to their significant increase in theTD group The absence in the ASD group of a significantincrease in vocalizations to respond is in agreement with thegeneral absence of an increase in the duration of the stateresponsiveness that we have described above

This result showing an atypical pattern of vocalizationdevelopment during the first year of life is comparable tothat of a prospective study by Ozonoff et al [32] on siblingswho later developed an ASD where it was found that at


0ndash6 6ndash12 0ndash6 6ndash12

Dur

atio

n (

)

5470 533

716

463

87

666

824769

Infantrsquos responsiveness ASDInfantrsquos responsiveness TD

Caregiverrsquos involvement ASDCaregiverrsquos involvement TD

(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6

ASD6ndash12 0ndash6 6ndash12

TD TD

Infantrsquos responsiveness

(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD

Caregiverrsquos involvement

(c)

Figure 1 Trajectories of infantrsquos responsiveness versus caregiverrsquos involvement (a) Trajectories (from T1 to T2) of infantrsquos responsiveness (b)and caregiverrsquos involvement (c) TD infants increase their responsiveness from T1 to T2 while in infants with ASD responsiveness remainsstable Caregivers of infants with ASD tend to decrease the time they spend in involvement (b) (c) Means and standard deviations (errorbars) lowastlowast119875 le 001 lowast119875 le 005

the age of six months there were no significant differencesin the frequency of either vocalizations or smiles comparedto TD infants (indeed the values of children with ASD wereslightly higher as in our sample) while in the second sixmonths a reduction in frequency occurred that resulted insignificant differences by the age of twelve months SimilarlyGoldberg et al [33] have reported that very often parentsrefer with a good level of accuracy their childrsquos regressionin the domain of early vocalizations These different resultsfor early vocalizations could also be interesting in the light ofGrossmanrsquos study [34] which provided evidence of a voice-sensitive region in the brain of 7-month-old but not 4-month-old infants Other studies as well as ours go in thesame direction as far as typical development is concernedin fact in TD infants an important increase in vocalizationswas found between the first and the second six months of lifewhereas the reduction in the total amount of vocalizations inthe ASD group could signify a failure in this developmentalstep The reduced use of vocalizations to respond couldsuggest that in infants with ASD the original vocalizations

do not develop into a vocal communication able to respondengage in dialogue evolving ultimately into structured lan-guage Congruently with the four to seven month perioddescribed as critical for the development of both language[35] and cerebral voice processing [34] the atypical patternof vocalization found in the ASD group highlights the periodbetween the first and the second six months as critical for thedevelopment of autism

As for caregiver behaviors the most significant resultwas the reduction between T1 and T2 of affectionate touchin the ASD group Affectionate touch refers to caregiverstouching the infant in an affectionate way (ie caressingor kissing himher) A natural and progressive decrease incaregiversrsquo affectionate touch during the first year of life oftypical children has been described by Ferber et al [36]We also found a decrease in this behavior in our TD grouphowever the decrease is less rapid and does not reach asignificant level as it does for caregivers of infants with ASDThe different trajectory of affectionate touch from T1 toT2 (Figure 3) is in agreement with the significant decrease


1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast

Total motor activity ASDTotal motor activity TD

Vocalizations ASDVocalizations TD

Rate

min

(a)

0

5

10

15

20

25

30

35

40

ASD ASD0ndash6 0ndash66ndash12 6ndash12

TD TD

lowast

lowast

Total motor activity

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)

Figure 2 Trajectories of infantrsquos total motor activity versus infantrsquos vocalizations (a) Trajectories (from T1 to T2) of total motor activity (b)and vocalizations total (c) Infants with ASD compared to TD infants show a reduced ratemin of motor activity at T1 and a reduced rateminof vocalizations at T2 (b) (c) Means and standard deviations (error bars) lowastlowast119875 le 001 lowast119875 le 005

from T1 to T2 in the duration of the state involvementWe could hypothesize that the decrease in involvement andaffectionate touch is probably related to the convergent actionof the atypical motor and vocal development during thefirst year of life of infants with ASD One hypothesis couldbe that caregivers decrease their use of touch to involvetheir infants as a result of the reduced responsiveness andattuned behaviors described in children with ASD Anotherhypothesis to take into account concerns the caregiverrsquosinherent social skills which are likely to be different acrosssubjects and especially across groups asASD is a dimensionaltrait that runs in families However more powerful analysesshould be used to validate these considerations In Figure 3it is also possible to observe the opposite trajectory ofthe caregiverrsquos stimulating gestures (caregiverrsquos gesticulatingtickling making faces or presenting objects to the infant)In fact caregivers of infants with ASD present an increasein stimulating gestures to involve their infants that is notpresent in the TD group Two of our previous studies onhome movies pointed to the increase in parental solicitations(both ldquoregulating uprdquo and ldquotouchingrdquo) as an early marker

for ASD [11 14] Moreover a recent paper on parent-infantinteraction in infant siblings at risk for autism has shownthat caregiversrsquo interactive behaviors are more directive [10]The current study adds another aspect to these previousfindings on the way parents of infants with ASD adapt totheir less responsive child whichmight appear contradictoryOn the one hand parents of infants who are developingan ASD might consider their childrenrsquos behavior to be areflection of hisher temperamental attitude and adapt theirbehavior by reducing affectionate touch on the other handthey might recognize that something is going wrong in theirtoo passive and unresponsive infant and congruently increasetheir stimulating attitude Nevertheless because the infantrsquosbehaviors are reduced in terms of quality but not in theduration of involvement and responsiveness parents mightnotice only some slight differences and slowly grow in theirawareness that their child needs further solicitation In factone methodological aspect regarding the study conducted bySaint-Georges et al [14] is crucial the use of computationalmethods permitted the selection of only successful interactivepatterns (meaning a caregiver solicitation followed by an


Rate

min

123

95 62

3

214

197

26

217

Affectionate touch ASDAffectionate touch TD

Stimulating gestures ASDStimulating gestures TD

0ndash6 0ndash66ndash12 6ndash12

(a)

0

5

10

15

20

25

30

35

40

ASD0ndash6 ASD6ndash12 0ndash6

TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD

Stimulating gestures

(c)

Figure 3 Trajectories of caregiverrsquos affectionate touch verses caregiverrsquos stimulating gestures (a) Trajectories (from T1 to T2) of affectionatetouch (b) and stimulating gestures (c) In the transition from T1 to T2 caregivers of infants with ASD compared to caregivers of TD infantsshow a significant reduction of affectionate touch whereas the ratemin of stimulating gestures increases (b) (c) Means and standarddeviations (error bars) lowastlowast119875 le 001 lowast119875 le 005

infant response) It is in this specific context that caregivertouching and regulating up is significantly increased inparents of infants with ASD compared to parents of TDinfants

5 Limitations

The first and most relevant limitation is represented bythe small sample size as the Type I error cannot be ruledout and the results may not survive multiple comparisontests Moreover we have described different patterns in thetwo groups by carrying out separate analyses of differencesbetween and within groups through Studentrsquos t-test butit was not possible to perform a formal statistical test ofthe interaction between these differences Given the smallsample size and the limited power of the statistical analysisresults should be considered a testable hypothesis for futureadequately powered studies

A second limitation is the quality of the videotapesrecorded by parents in a naive context which varies acrosssubjects and reduces the comparability of the sequences interms of opportunity for reciprocal exchanges To reduce thebias of the wide variability of video sequences we chose

situations of interaction between infant and caregiver assimilar to each other as possible selecting the video materialspecifically on the basis of the scenes which were filmed

Finally a further limitation is the absence of a controlgroup consisting of infants with non-ASDneurodevelopmen-tal disorders (eg intellectual disability) In our study ASDandTDgroupswere notmatched based on IQ thuswe cannotexclude the possibility that the differences are in part dueto this factor Therefore we are aware that this study lacksevidence for the specificity of the abnormalities found ininfants with ASD

6 Conclusions

To summarize the present study suggests that infants withearly onset autism spectrum disorders may display bothmotor and vocal impairments We hypothesize that a non-synchronic motor-vocal pattern may interfere in differentways with the development of reciprocity in the primaryrelationship between infants later diagnosed with ASD andtheir caregivers Differences in motor activity patterns areevident by the end of the first six months of life and wesuggest that theymay have a role in the lack of initiative and in


difficulties in having an active role in purposeful interactionsas well as in reciprocal responses On the other hand areduction of vocalizations across time and in comparisonwith TD infants may be seen as a red flag of atypical socialdevelopment

Given the limitations these findings should be regardedas very preliminary encouraging further research in thisdirection with a dual purpose to evaluate if the differenceswe have highlighted withstand more powerful statisticalanalysis by increasing the sample size and to determine thespecificity of the results by introducing a control group ofinfants with non-ASD developmental disorders Moreoverfuture research addressing the topic of infantsrsquo motor-vocaldevelopment might benefit from a prospective approachwhich would enable the use of standardized observations ofinfants at risk

If this nonsynchronic motor-vocal pattern observed ininfants later diagnosed with ASD were to be confirmed byfuture studies its early identification might provide usefulinformation for determining infants at risk for ASD andstarting a timely intervention

References

[1] B J Anderson PVietze andP RDokecki ldquoReciprocity in vocalinteractions ofmothers and infantsrdquoChildDevelopment vol 48no 4 pp 1676ndash1681 1977

[2] T Cole and J B Teboul ldquoNon-zero-sum collaboration reci-procity and the preference for similarity developing an adaptivemodel of close relational functioningrdquo Personal Relationshipsvol 11 no 2 pp 135ndash160 2004

[3] T B van Ommeren S Begeer A M Scheeren and H MKoot ldquoMeasuring reciprocity in high functioning children andadolescents with autism spectrum disordersrdquo Journal of Autismand Developmental Disorders pp 1001ndash1010 2012

[4] M Carpenter ldquoJust how joint is joint action in infancyrdquo Topicsin Cognitive Science vol 1 no 2 pp 380ndash392 2009

[5] R Feldman and C W Greenbaum ldquoAffect regulation and syn-chrony in mother-infant play as precursors to the developmentof symbolic competencerdquo Infant Mental Health Journal vol 18no 1 pp 4ndash23 1997

[6] C Trevarthen and K J Aitken ldquoInfant intersubjectivityresearch theory and clinical applicationsrdquo Journal of ChildPsychology and Psychiatry and Allied Disciplines vol 42 no 1pp 3ndash48 2001

[7] D N SternThe First Relationship Infant and Mother HarvardUniversity Press Cambridge UK 1977

[8] F Muratori and S Maestro ldquoAutism as a downstream effect ofprimary difficulties in intersubjectivity interacting with abnor-mal development of brain connectivityrdquo International Journalfor Dialogical Science vol 2 no 1 pp 93ndash118 2007

[9] R Palomo M Belinchon and S Ozonoff ldquoAutism and familyhomemovies a comprehensive reviewrdquo Journal of Developmen-tal and Behavioral Pediatrics vol 27 no 2 supplement pp S59ndashS68 2006

[10] M W Wan J Green M Elsabbagh et al ldquoParent-infantinteraction in infant siblings at risk of autismrdquo Research inDevelopmental Disabilities vol 33 no 3 pp 924ndash932 2012

[11] F Muratori F Apicella P Muratori and S Maestro ldquoInter-subjective disruptions and caregiver-infant interaction in early

autistic disorderrdquo Research in Autism SpectrumDisorders vol 5no 1 pp 408ndash417 2011

[12] S M Chow J D Haltigan and D S Messinger ldquoDynamicinfant-parent affect coupling during the face-to-facestill-facerdquoEmotion vol 10 no 1 pp 101ndash114 2010

[13] C Saint-Georges R S Cassel D Cohen et al ldquoWhat studiesof family home movies can teach us about autistic infants aliterature reviewrdquo Research in Autism Spectrum Disorders vol4 no 3 pp 355ndash366 2010

[14] C Saint-Georges A Mahdhaoui M Chetouani et al ldquoDo par-ents recognize autistic deviant behavior long before diagnosisTaking into account interaction using computational methodsrdquoPLoS One vol 6 no 7 article e22393 2011

[15] J A Doussard-Roosevelt C M Joe O V Bazhenova and SW Porges ldquoMother-child interaction in autistic and nonautisticchildren characteristics of maternal approach behaviors andchild social responsesrdquo Development and Psychopathology vol15 no 2 pp 277ndash295 2003

[16] C Trevarthen and S Daniel ldquoDisorganized rhythm and syn-chrony early signs of autism and Rett syndromerdquo Brain andDevelopment vol 27 supplement 1 pp S25ndashS34 2005

[17] M H van Ijzendoorn A H Rutgers M J Bakermans-Kranenburg et al ldquoParental sensitivity and attachment in chil-dren with autism spectrum disorder comparison with childrenwith mental retardation with language delays and with typicaldevelopmentrdquo Child Development vol 78 no 2 pp 597ndash6082007

[18] E A Thomas and J A Martin ldquoAnalyses of parent-infantinteractionrdquo Psychological Review vol 83 no 2 pp 141ndash1561976

[19] American Psychiatric Association Diagnostic and StatisticalManual ofMental Disorders American Psychiatric PressWash-ington DC USA 4th edition 2000

[20] T M Achenbach and L Rescorla Manual for the ASEBA Pre-school Forms and Profile ASEBA Burlington VT USA 2000

[21] S Maestro F Muratori M C Cavallaro et al ldquoHow youngchildren treat objects and people an empirical study of thefirst year of life in autismrdquo Child Psychiatry and HumanDevelopment vol 35 no 4 pp 383ndash396 2005

[22] L Zwaigenbaum S Bryson T Rogers W Roberts J Brianand P Szatmari ldquoBehavioral manifestations of autism in thefirst year of liferdquo International Journal of Developmental Neu-roscience vol 23 no 2-3 pp 143ndash152 2005

[23] Z Biringen ldquoEmotional availability conceptualization andresearch findingsrdquo The American Journal of Orthopsychiatryvol 70 no 1 pp 104ndash114 2000

[24] NoldusThe Observer Reference Manual Version 10 0 2010[25] SPSS Inc SPSS For Windows Version 16 0 SPSS Inc Chicago

Ill USA 2007[26] J L Adrien P Lenoir J Martineau et al ldquoBlind ratings of early

symptoms of autism based upon family home moviesrdquo Journalof theAmericanAcademy of Child andAdolescent Psychiatry vol32 no 3 pp 617ndash626 1993

[27] G T Baranek ldquoAutism during infancy a retrospective videoanalysis of sensory-motor and social behaviors at 9ndash12 monthsof agerdquo Journal of Autism and Developmental Disorders vol 29no 3 pp 213ndash224 1999

[28] H Phagava F Muratori C Einspieler et al ldquoGeneral move-ments in infants with autism spectrum disordersrdquo GeorgianMedical News no 156 pp 100ndash105 2008


[29] V Guinchat B Chamak B Bonniau et al ldquoVery early signs ofautism reported by parents include many concerns not specificto autism criteriardquo Research in Autism Spectrum Disorders vol6 no 2 pp 589ndash601 2012

[30] J E Flanagan R Landa A Bhat and M Bauman ldquoHead lag ininfants at risk for autism a preliminary studyrdquo The AmericanJournal of Occupational Therapy vol 66 no 5 pp 577ndash5852012

[31] K A Fournier C J Hass S K Naik N Lodha and J HCauraugh ldquoMotor coordination in autism spectrum disordersa synthesis and meta-analysisrdquo Journal of Autism and Develop-mental Disorders vol 40 no 10 pp 1227ndash1240 2010

[32] S Ozonoff A M Iosif F Baguio et al ldquoA prospective study ofthe emergence of early behavioral signs of autismrdquo Journal of theAmerican Academy of Child and Adolescent Psychiatry vol 49no 3 pp 256e1-2ndash266e1-2 2010

[33] W A Goldberg K L Thorsen K Osann and M A SpenceldquoUse of home videotapes to confirm parental reports of regres-sion in autismrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1136ndash1146 2008

[34] T Grossmann R Oberecker S P Koch and A D FriedericildquoThe developmental origins of voice processing in the humanbrainrdquo Neuron vol 65 no 6 pp 852ndash858 2010

[35] P K Kuhl ldquoEarly language acquisition cracking the speechcoderdquo Nature Reviews Neuroscience vol 5 no 11 pp 831ndash8432004

[36] S G Ferber R Feldman and I R Makhoul ldquoThe developmentof maternal touch across the first year of liferdquo Early HumanDevelopment vol 84 no 6 pp 363ndash370 2008


Palomo et al [9] suggested that in infants with ASDcompared to TD infants there are some differences in theability to interact reciprocally because of a reduction inthe capacity of the former to engage with the other andto orient toward social stimuli Wan et al [10] found thatinfants at-risk of ASD showed lower levels of liveliness thatis a general tendency toward less activity during exchangesWe have described a general difficulty in maintaining socialengagement and in accepting invitation [11] Other studies[12 13] have revealed that during mutual exchange betweeninfant and parent while the former serves as the leadingmember in a typical dyad this does not happen in dyads withan infant with ASD Infants with these social difficulties offertheir parents who are in need of both responses and promptsfrom their infants fewer opportunities for interaction Whileinfantsrsquo social difficulties have been sufficiently analyzedand assessed their effects on parental attitude have receivedless attention Some studies have described longitudinalchanges in the intensity and quality of parental solicitationsSaint-Georges et al [14] focused on sequentially analyzedinteractions and they found that parents of infants with ASD(1) respond to their infants at the same level as parents ofinfants with TD but (2) show a more directive interactivestyle characterized by an increase in stimulating and touchingactivity Congruently Doussard-Roosevelt et al [15] andTrevarthen and Daniel [16] found that parents of childrenwith ASD overstimulate their infants as a result of theirinactivity Van Ijzendoorn et al [17] observed that parentsof children with ASD tended to answer their childrsquos signalswith more physical contact and fewer verbal approaches Wehave described [11] a reduction in caregiver behaviors aimedat downregulating the infantrsquos arousal and mood as one ofthe first expressions of lack of initiative in infants with ASD

This evidence of an early alteration in parent-infantinteraction suggests that the first signs of ASD could be betterunderstood by observing interactive dynamics rather thanthe increase or decrease in single behaviors of one memberof the dyad As stated by Thomas and Martin [18] and byAnderson et al [1] parent-infant reciprocity is better studiedby analyzing the effects on the other member of each dyadmemberrsquos behavior

On the basis of these assumptions we have hypothesizedthat reciprocity can be better understood by analyzing dyadicbehaviors that occur in simple caregiver-infant interactionsobserved in family home videos a valid method for observ-ing ecological interactions from the first stages of life Todescribe interactions specifically in terms of reciprocity wehave developed the caregiver-infant reciprocity scale (CIRS)which consists of items referring to mutual exchanges (suchas involvement or responsiveness) and single or molecularbehaviors (such as vocalization for infants or name promptfor caregivers) The CIRS was developed in order to describeand evaluate how infants try to involve and to respond to theotherrsquos bids according to the literature [9 11] we expectedin infants with ASD a reduced frequency of behaviors usedto involve and to maintain the other in interaction At thesame time the CIRS enabled us to explore the effect of theinfantrsquos behavior on parentsrsquo interactive attitudes accordingto the literature we expected some differences in terms of

type and quantity of stimulations but not in terms of globalavailability [17] The CIRS was also developed to explorewhether and to what extent abnormalities in the frequencyof motor behaviors contribute to disturbing the developmentof the atypical infantrsquos ability to engage in interactions ourhypothesis based on a previous exploratory study was thatabnormalities in motor activity can constitute a warningsignal for difficulties in reciprocity

2 Methods

21 Participants We studied retrospective home moviesfrom the first year of life of two groups of children Thefirst group was composed of 10 children (MF 91) with adiagnosis of autism spectrum disorder (ASD) performed bya child psychiatrist according to the DSM-IV-TR criteria [19]and confirmed by ADOS-G and ADI-R scores At the time ofthe diagnosis their ages ranged between 4 and 6 years Theywere also administered Griffithsmental developmental scale-ER in order to determine intellectual functioning (meanIQ 5926 sd 849) The second group was composed of9 typically developing (TD) children who were recruitedamong children attending a local kindergarten these childrendid not present any abnormal medical or developmentalcondition as confirmed by nonclinical scores using the childbehavior checklist [20] At the time of recruitment theTD children were aged between 4 and 5 years The videosused in the study were taken from a larger collection ofvideos held at the Stella Maris Institute University of PisaItaly

22 Video Collection and Editing Procedures Families wereasked to provide any videotape recorded during their childrsquosfirst year of life Copies of the videos were made and codedwith an ID number to preserve confidentiality Writteninformed consent was collected from participating familiesThe selection of the video clips was carried out by a researchassistant naive to the childrenrsquos diagnoses Only sequenceswhere the child was involved in an interaction with avisible caregiver were selected for the study Sequences wereselected in terms of situation type We included face-to-faceinteractions in various circumstances such as bath timenutrition time and play time We also included face-to-faceinteractions per se (ie the baby on the changing table and themother in front of himher talking in motherese) Birthdayparties and sequences with several adults andor childrensimultaneously present in the video were excluded

A t-test was performed to check that the selectedmaterialwas comparable across groups in terms of total durationand number and type of situations In accordance withother studies in the literature [11 21 22] which analyzedthe developmental pattern of infants across the first year oflife we considered the first and second six months of lifeseparately in order to take into account the multiple changeswhich take place during this period The total duration ofthe videos analyzed was 46 minutes for the age period 0ndash6months (hereinafter T1) and 51 minutes for the age period6ndash12 months (hereinafter T2) We analyzed a total of 81



Infant behaviors














Table 1 Continued
























3 Results







4 Discussion








ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

















Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References









































Infant behaviors














Table 1 Continued
























3 Results







4 Discussion








ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

















Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References








































Table 1 Continued
























3 Results







4 Discussion








ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

















Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References














































3 Results







4 Discussion








ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

















Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References













































ratemin 22 (17) 24 (16) 16 (07) 2 (08) 077 036 038 039Events(5)

















Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References













































Events(5)













Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References









































Dur

atio

n (

)

5470 533

716

463

87

666

824769



(a)

0

20

40

60

80

100

120

lowast

ASD0ndash6


TD TD


(b)

0

20

40

60

80

100

120lowast

ASD0ndash6


TD TD


(c)






1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References








































1 2 3 4

98

59

184

109

lowast

156

143

101

255

lowastlowast



Rate

min

(a)

0

5

10

15

20

25

30

35

40


TD TD

lowast

lowast


(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD

6ndash12TD

lowast

lowast

Vocalizations

(c)





Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References








































Rate

min

123

95 62

3

214

197

26

217




(a)

0

5

10

15

20

25

30

35

40


TD6ndash12TD

lowast

Affectionate touch

(b)

0

5

10

15

20

25

30

35

40

ASD0ndash6

ASD6ndash12 0ndash6

TD6ndash12TD


(c)



5 Limitations





6 Conclusions






References











































References





































